Partial proportional odds model - An alternate choice for analyzing pedestrian crash injury severities

Lekshmi Sasidharan, Monica Menendez

    Research output: Contribution to journalArticle

    Abstract

    The conventional methods for crash injury severity analyses include either treating the severity data as ordered (e.g. ordered logit/probit models) or non-ordered (e.g. multinomial models). The ordered models require the data to meet proportional odds assumption, according to which the predictors can only have the same effect on different levels of the dependent variable, which is often not the case with crash injury severities. On the other hand, non-ordered analyses completely ignore the inherent hierarchical nature of crash injury severities. Therefore, treating the crash severity data as either ordered or non-ordered results in violating some of the key principles. To address these concerns, this paper explores the application of a partial proportional odds (PPO) model to bridge the gap between ordered and non-ordered severity modeling frameworks. The PPO model allows the covariates that meet the proportional odds assumption to affect different crash severity levels with the same magnitude; whereas the covariates that do not meet the proportional odds assumption can have different effects on different severity levels. This study is based on a five-year (2008-2012) national pedestrian safety dataset for Switzerland. A comparison between the application of PPO models, ordered logit models, and multinomial logit models for pedestrian injury severity evaluation is also included here. The study shows that PPO models outperform the other models considered based on different evaluation criteria. Hence, it is a viable method for analyzing pedestrian crash injury severities.

    Original languageEnglish (US)
    Pages (from-to)330-340
    Number of pages11
    JournalAccident Analysis and Prevention
    Volume72
    DOIs
    StatePublished - Jan 1 2014

    Fingerprint

    pedestrian
    Wounds and Injuries
    Logistic Models
    Switzerland
    Pedestrian safety
    Pedestrians
    Safety
    evaluation

    Keywords

    • Comparison
    • Crash severity
    • Multinomial logit model
    • Ordered logit model
    • Partial proportional odds model
    • Pedestrian

    ASJC Scopus subject areas

    • Human Factors and Ergonomics
    • Safety, Risk, Reliability and Quality
    • Public Health, Environmental and Occupational Health
    • Law

    Cite this

    Partial proportional odds model - An alternate choice for analyzing pedestrian crash injury severities. / Sasidharan, Lekshmi; Menendez, Monica.

    In: Accident Analysis and Prevention, Vol. 72, 01.01.2014, p. 330-340.

    Research output: Contribution to journalArticle

    @article{cda78146854248eeaaed61c357abceca,
    title = "Partial proportional odds model - An alternate choice for analyzing pedestrian crash injury severities",
    abstract = "The conventional methods for crash injury severity analyses include either treating the severity data as ordered (e.g. ordered logit/probit models) or non-ordered (e.g. multinomial models). The ordered models require the data to meet proportional odds assumption, according to which the predictors can only have the same effect on different levels of the dependent variable, which is often not the case with crash injury severities. On the other hand, non-ordered analyses completely ignore the inherent hierarchical nature of crash injury severities. Therefore, treating the crash severity data as either ordered or non-ordered results in violating some of the key principles. To address these concerns, this paper explores the application of a partial proportional odds (PPO) model to bridge the gap between ordered and non-ordered severity modeling frameworks. The PPO model allows the covariates that meet the proportional odds assumption to affect different crash severity levels with the same magnitude; whereas the covariates that do not meet the proportional odds assumption can have different effects on different severity levels. This study is based on a five-year (2008-2012) national pedestrian safety dataset for Switzerland. A comparison between the application of PPO models, ordered logit models, and multinomial logit models for pedestrian injury severity evaluation is also included here. The study shows that PPO models outperform the other models considered based on different evaluation criteria. Hence, it is a viable method for analyzing pedestrian crash injury severities.",
    keywords = "Comparison, Crash severity, Multinomial logit model, Ordered logit model, Partial proportional odds model, Pedestrian",
    author = "Lekshmi Sasidharan and Monica Menendez",
    year = "2014",
    month = "1",
    day = "1",
    doi = "10.1016/j.aap.2014.07.025",
    language = "English (US)",
    volume = "72",
    pages = "330--340",
    journal = "Accident Analysis and Prevention",
    issn = "0001-4575",
    publisher = "Elsevier Limited",

    }

    TY - JOUR

    T1 - Partial proportional odds model - An alternate choice for analyzing pedestrian crash injury severities

    AU - Sasidharan, Lekshmi

    AU - Menendez, Monica

    PY - 2014/1/1

    Y1 - 2014/1/1

    N2 - The conventional methods for crash injury severity analyses include either treating the severity data as ordered (e.g. ordered logit/probit models) or non-ordered (e.g. multinomial models). The ordered models require the data to meet proportional odds assumption, according to which the predictors can only have the same effect on different levels of the dependent variable, which is often not the case with crash injury severities. On the other hand, non-ordered analyses completely ignore the inherent hierarchical nature of crash injury severities. Therefore, treating the crash severity data as either ordered or non-ordered results in violating some of the key principles. To address these concerns, this paper explores the application of a partial proportional odds (PPO) model to bridge the gap between ordered and non-ordered severity modeling frameworks. The PPO model allows the covariates that meet the proportional odds assumption to affect different crash severity levels with the same magnitude; whereas the covariates that do not meet the proportional odds assumption can have different effects on different severity levels. This study is based on a five-year (2008-2012) national pedestrian safety dataset for Switzerland. A comparison between the application of PPO models, ordered logit models, and multinomial logit models for pedestrian injury severity evaluation is also included here. The study shows that PPO models outperform the other models considered based on different evaluation criteria. Hence, it is a viable method for analyzing pedestrian crash injury severities.

    AB - The conventional methods for crash injury severity analyses include either treating the severity data as ordered (e.g. ordered logit/probit models) or non-ordered (e.g. multinomial models). The ordered models require the data to meet proportional odds assumption, according to which the predictors can only have the same effect on different levels of the dependent variable, which is often not the case with crash injury severities. On the other hand, non-ordered analyses completely ignore the inherent hierarchical nature of crash injury severities. Therefore, treating the crash severity data as either ordered or non-ordered results in violating some of the key principles. To address these concerns, this paper explores the application of a partial proportional odds (PPO) model to bridge the gap between ordered and non-ordered severity modeling frameworks. The PPO model allows the covariates that meet the proportional odds assumption to affect different crash severity levels with the same magnitude; whereas the covariates that do not meet the proportional odds assumption can have different effects on different severity levels. This study is based on a five-year (2008-2012) national pedestrian safety dataset for Switzerland. A comparison between the application of PPO models, ordered logit models, and multinomial logit models for pedestrian injury severity evaluation is also included here. The study shows that PPO models outperform the other models considered based on different evaluation criteria. Hence, it is a viable method for analyzing pedestrian crash injury severities.

    KW - Comparison

    KW - Crash severity

    KW - Multinomial logit model

    KW - Ordered logit model

    KW - Partial proportional odds model

    KW - Pedestrian

    UR - http://www.scopus.com/inward/record.url?scp=84907341854&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84907341854&partnerID=8YFLogxK

    U2 - 10.1016/j.aap.2014.07.025

    DO - 10.1016/j.aap.2014.07.025

    M3 - Article

    C2 - 25113015

    AN - SCOPUS:84907341854

    VL - 72

    SP - 330

    EP - 340

    JO - Accident Analysis and Prevention

    JF - Accident Analysis and Prevention

    SN - 0001-4575

    ER -